In contrast to short-term kidney allograft survival, long-term graft survival has not improved dramatically. The leading cause of late graft failure in living recipients is chronic allograft nephropathy (CAN), accounting for 35-40% of graft failure 6 months post-transplant. The precise etiology of CAN is not known, but immunologic factors are thought to be critical: 1) CAN is associated with acute rejection and , 2) living related allografts show improved graft survival compared to cadaveric grafts. Alternatively, immunosuppressive therapy's lack of effect on CAN suggests nonimmunologic factors may also be involved. Factors often mentioned include size donor-recipient size disparity, drug nephrotoxicity, hypertension, and hyperlipidemia. In CAN, combined immunologic and non-immunologic responses to the allograft result in tissue damage, inflammatory cell infiltration, and local fibroblast proliferation with the end result of interstitial matrix deposition and fibrosis. The process appears to be regulated by cytokines and growth factors such as transforming growth factor-beta (TGF-beta), basic fibroblast growth factor (bFGF), vascular endothelial growth factor (VEGF), and platelet derived growth factor (PDGF). While investigators have utilized a number of rat models to study CAN, these models of vascularized transplants have utilized donor/recipient strain combinations with only minor antigen differences and pathologic changes occur 4-6 months post-transplant making the models labor-intensive and inefficient. We recently observed that while mouse kidney allografts survive for prolonged periods without immunosuppression (1;2) that the grafts' function deteriorates gradually and with histologic features typical of human CAN by 6 weeks post transplant (3). Further, consistent with CAN, we found that 1) TGF-beta was highly expressed in chronically rejecting allografts compared to nonrejecting isografts, and 2) enhanced kidney matrix deposition (3). Despite reducing the donor/recipient MHC disparity, with an associated reduction in intra-graft TGF-beta expression, CAN severity was unchanged. Further, TGF-beta, while itself a potent fibrogenic cytokine, also has immunosuppressive qualities believed to be beneficial to graft survival making it less desirable as a target to prevent CAN. These studies suggest that while MHC disparity may regulate TGF-beta and CAN development, other local factors influence CAN. For instance, we found that alloantibody development is critical for CAN development using the mouse model(4). We've proposed that reducing extra-cellular matrix deposition or organization may affect the CAN outcome. Our strategy is to abrogate matrix production directly using pharmacologic inhibitors of essential matrix synthesis enzymes. Prolyl-4-hydroxylase is an enzyme essential for collagen formation as it post-translationally modifies the procollagen alpha chains. Prolyl-4-hydroxylase inhibition prevents proline hydroxylation of procollagen chains leading to unstable procollagen folding. The newly formed abnormal procollagen is degraded thus decreasing interstitial collagen deposition. Novel phenanthrolinone compounds, competitive inhibitors of prolyl 4-hydroxylase both in vivo and in vitro, do not appear to have systemic toxicity. We studied the effects of a novel, proprietary, orally active phenanthrolinone prolyl-4 hydroxylase (PHI) inhibitor in our mouse CAN model (5). Kidneys from C57BL/6 (H-2b) mice were transplanted into MHC-incompatible CBY (H-2d) recipients (allografts) or, as a non-rejecting control group, their littermates. At 3 weeks post-transplant, allografted mice received PHI (50mg/kg per day) or vehicle for 3 weeks, and we assayed for renal function and histomorphologic changes associated with CAN 6 weeks post-transplant. PHI treatment was well tolerated; treated mice had stable body weights and displayed no apparent skeletal abnormalities. Most important, the glomerular filtration rate was significantly greater in PHI treated mice with allografts (3.3 +/- 0.5 ml/min/kg) compared to those receiving vehicle (1.8 +/- 0.5 ml/min/kg, p<0.05), while renal function was unimpaired in mice with isografts (6.45 +/- 0.53 ml/min/kg). While we saw no difference in vascular and glomerular injury severity between groups, the allografts displayed CAN histologic changes but the severity was reduced in those receiving PHI (12.6 +/- 0.8) compared to vehicle (15.4 +/- 1.2; p<0.05), with reduced allograft interstitial inflammation and fibrosis. PHI treatment was also associated with a substantial reduction in urinary protein (0.6 +/- 0.1 mg protein/mg creatinine) compared to vehicle treated recipients (1.0 +/- 0.3 mg protein/mg creatinine). To study how PHI might mediate these effects, we analyzed intragraft cytokine profiles using real-time PCR and, with the exception of IL-4, found no significant differences for Th1 or Th2 cytokines, nor for TGF-beta, between allograft groups. In PHI-treated recipients however, there was a significant reduction in relative mRNA IL-4 expression (6.12-fold expression relative to normal kidney) compared to vehicle treated allografts (12.86-fold relative to normal kidney; p=0.04). Thus, despite PHI significantly reducing the inflammatory cell infiltrate within allografts, we observed only modest intra-graft cytokine profile changes. These studies demonstrate that PHI treatment can reduce matrix deposition in CAN and may represent a novel therapy for humans. In other studies, we've pursued CTGF, a 38 Kd cysteine rich protein and a member of the CCN growth factor family, that is produced by kidney endothelial cells, fibroblasts, and mesangial cells. CTGF stimulates the production of collagen and fibronectin matrix proteins. Further, while CTGF expression is regulated by TGF-beta, it lacks TGF-beta's immunoregulatory properties. CTGF is upregulated in a number of rodent renal disease models and is implicated in the fibrotic process. In human renal diseases associated with fibrosis, CTGF expression is also enhanced. We studied CTGF as a target downstream of TGF-beta but one apparantly involved in matrix accumulation as a potential therapeutic approach to prevent or treat CAN. We measured CTGF expression using real time RT-PCR (confirmed by Northern) and found CTGF mRNA expression significantly increased (2.7-fold) in mouse kidney allografts with CAN compared to nonrejecting kidneys 6 weeks after transplant (6). CTGF mRNA expression was preceded by a nearly 12-fold and 6-fold TGF-beta mRNA induction in allografts at one week and two weeks post-transplant, respectively, compared to naive kidneys. Thus CTGF mRNA is expressed in allografts with CAN, and is associated with a rise in TGF-beta mRNA. We also examined CTGF in a pilot study of NIH human kidney transplant recipients (6). Mean serum (50.1 +/- 3.9 ng/ml; n=25) and urine CTGF levels (32.2 +/- 6.0 ng/mg creatinine; n=20) were significantly elevated compared to normal, healthy individuals (n=10; 15.4 +/- 3.6 ng/ml and 0.8 +/- 0.2 ng/mg creatinine, respectively). While recipient factors like age, time post transplant, calcineurin inhibitor use, and cold ischemic time did not correlate to CTGF levels, mean urinary CTGF levels were highest in patients with biopsies demonstrating CAN, and lowest in rerecipients with normal histology. The goal of these studies is to develop novel interventions and screening methods for transplant recipients at risk for developing chronic rejection. REFERENCES (1)Transplantation 1995; 59:746-755. (2)Transplantation 2000; 69:2137-2143. (3)Kidney Int 1999; 55: 1935-1944. (4)Kidney Int 2002; 62: 290-300. (5)Fransceschini N, Cheng O, Zhang X, Ruiz P, Mannon RB. Prolyl-hydroxylase inhibition ameliorates chronic allograft nephropathy in the mouse. Submitted. (6)J Am Soc Nephrol 2002; 13:24A